Development and validation of a GC-MS method for determination of metformin in normal brain and in glioblastoma tissues.

Metformin hydrochloride (MH) has recently been repurposed as an anticancer agent, showing antiproliferative activity in vitro and in vivo. In particular, experimental evidence has suggested its potential clinical efficacy in glioblastoma (GBM), a very aggressive tumor frequently characterized by gloomy prognosis. Unfortunately, the published literature concerning experimental applications of MH in glioblastoma animal models report no data on metformin levels reached in the brain, which, considering the high hydrophilicity of the drug, are likely very low. Therefore, new sensitive analytical methods to be applied on biological tissues are necessary to improve our knowledge of MH in vivo biodistribution and biological effects on tumors. In this research work, a GC-MS method for MH quantification in brain tissues is proposed. MH has been derivatized using N-methyl-bis(trifluoroacetamide), as already described in the literature, but the derivatization conditions have been optimized; moreover, deuterated MH has been selected as the best internal standard, after a comparative evaluation including other internal standards employed in published methods. After ascertaining method linearity, its accuracy, precision, specificity, repeatability, LOD and LOQ (0.373 µM and 1.242 µM, respectively, corresponding to 0.887 and 2.958 pmol/mg of wet tissue) have been evaluated on mouse brain tissue samples, obtained through a straightforward preparation procedure involving methanolic extraction from lyophilized brain homogenates and solid phase purification. The method has been validated on brain samples obtained from mice, either healthy or xenografted with GBM cells, receiving metformin dissolved in the drinking water. This analytical method can be usefully applied in preclinical studies aiming at clarifying MH mechanism of action in brain tumors.